366 THE ILLUMINATION OP THE ECLIPSED MOON. 



eartli) than those we have been considering. This con- 

 clusion will bo perfectly obvious when it is remembered that 

 the refractive power of the atmosphere diminishes as its 

 rarity increases. If a line wore drawn in the diagram 

 representing the extreme range within which rays passing 

 through the lowest stratum shed their light along the axis 

 of the shadow, and similar linos were drawn for rays passing 

 at heights in the atmosphere of half a mile, a mile, two 

 miles, etc., these lines would overlap each other successively, 

 the result being that an observer at any one spot (except 

 jnst at the near end) would receive rays which had passed 

 through an infinite number of atmospheric strata, derived 

 from an infinite number of circles on the sun's disc. 



We are now prepared to examine what the state of things 

 will be at that spot in the shadow where the moon may 

 happen to be placed. The two extreme positions of tho 

 moon aro a(;curatcly shown in the diagr.am, and it will bo 

 soen how largely these positions differ in respect of their 

 distance from the earth. Tho vai-iation in tho apparent size 

 of the moon at different times must have boon often observed 

 by persons unacquainted with the cause. In truth, the 

 moon is one-seventh part further off in apogee than in 

 perigee, and its apparent diameter is therefore one-seventh 

 part greater in perigee than in apogee. The moon eclipsed 

 in either position will receive rays which have passed 

 through the lowest stratum of tho earth's atmosphere, 

 although it will be beyond tho spot whore what I have 

 called direct refraction through the lowest stratum ceases 

 to take effect. Apart, however, from the rays received 

 through tho lowest stratum (which woiild probably give very 

 little light), the moon in both positions will receive rays 

 refracted through the higher regions of the atmosphere, hut 



